Standing stem residues affect erosion, hydrology, and other processes differently than flat residues, but stem persistence under no-tillage management is not well understood. We developed an equation to predict standing stem number over time, based on precipitation and air temperature. Crops were field-grown winter and spring wheat (Triticum aestivum L.), barley (Hordeum vulgare L.), and oat (Avena sativa L.) grown near Bushland, TX, on Pullman clay loam (fine, mixed, thermic Torrertic Paleustoll). Fallow-period irrigation treatments produced three decomposition environments. Standing stems were counted in flagged quadrats 18, 98, 158, 223, 289, and 379 d after harvest. The daily minimum of precipitation-based moisture or mean air temperature coefficients was accumulated as decomposition days (DD). Standing stem fraction (SF) was predicted assuming SF = exp (k(DD − B)}. The threshold, B, was ≈17.5 DD for all crops, and k was −0.284, −0.176, −0.169, and −0.116 for oat, barley, and winter and spring wheat, respectively. Equation evaluation used data from North Dakota, Oregon, and Texas. Stem number prediction tended to be high before the B threshold and low later. Paired t-tests indicated no significant difference between predicted and measured stem fraction of spring wheat or barley. Stem fraction was overestimated by 0.09 for winter wheat averaged across Oregon and Texas data. Use of DDs improved prediction of standing stem persistence across diverse climates. Such information is needed for a wide range of erosion, water balance, and micrometeorological studies. A quantitative index for forces such as strong winds, animal traffic, or blowing precipitation may improve the model.